METALLURGICAL REPORT

Date: 9/13/04

Prepared for: Don Fogg Custom Knives

Background

Don Fogg Custom Knives is an internationally known name in high-quality custom knives and swords. The products are often manufactured using centuries-old techniques applied to modern steels. A persistent question among bladesmiths has been why a “Samurai” sword blade, which has been clayed in the traditional manner to harden only the cutting edge, bends down when quenched in oil and up when quenched in water. This investigation has been conducted to 1) evaluate the characteristic microstructures in both oil-quenched and water-quenched blades and 2) try to explain the aforementioned bending phenomenon.

Heat Treatment

It has been reported that the blades were forged from a 1.25” round bar of 1095 steel. The blades were normalized after forging and again after rough grinding. The blades were heated to 1415F and held at temperature until the color was uniform. The oil-quenched blade was quenched horizontally in Tough Quench oil at 85F and held until it stopped bubbling. The water quenched blade was heated to 1415F and held for the same amount of time and then quenched horizontally in water at 85F and held for a three second count, removed, and immediately quenched in oil until all activity stopped.

After removing the clay, the blades were tempered at 415F for one hour, removed and air cooled and tempered at 415F for another hour.

Samples As Received

Fig. 1 Blade samples as received. Top blade was water quenched, bottom blade was oil quenched. Both blades were made from a single piece of 1095 steel. Hamon line is visible at the arrows. Dotted lines indicate where samples for microstructural analysis were taken.

Laboratory Analysis

Each blade was sectioned using an abrasive saw and coolant flood to prevent altering the steel structure during cutting. Each section was marked on the spine section with a file. One groove indicated the oil quench and two grooves indicated the water quench. Both samples were mounted in a cold-mount medium (see Fig. 2 and Fig. 3).

Fig. 2. Cross section of oil-quenched Fig.3 Cross section of water-quenched

Blade. Blade.

In figures 2 and 3, the hamon line can be seen at the arrows.

After mounting and etching, the samples were analyzed using a Zeiss AxioMat Metallurgical microscope. All images are bright-field. All samples have been etched with 5% nitric acid in alcohol.

Fig. 4 Cutting edge, oil quench, 1000X. Fig. 5 Cutting edge, 1000X, water quench. Structure is tempered martensite and small Structure is tempered martensite with small

islands of pearlite or bainite (dark spots). islands of pearlite or bainite (dark spots).

Fig. 6 Hamon line, arrow, 50X oil quench. Fig. 7 Hamon line, water quench, 50X. Lines on

right edge are scratches. Triangular spot in

center is polishing imperfection.

Fig. 8 Hamon Line, 1000X, oil quench. Fig. 9 Hamon Line, 1000X, water quench.

Structure is fine pearlite or bainite (dark) Structure is tempered martensite and fine

Tempered martensite (light) and cementite pearlite or bainite (dark) and cementite

(white spots). (white spots).

Fig. 10 Spine, 1000X, oil quench. Structure Fig. 11 Spine, 1000X, water quench.

is fine pearlite, possibly martensite, and Structure is fine pearlite possibly martensite and

cementite (white spots). cementite.

Chemistry

A sample was cut and subjected to emission spectroscopy to determine the actual chemistry. It should be noted that the lab had no high carbon standard with which to calibrate the spectrometer and the carbon value shown may be slightly low.

C Mn P S Si

.81 .34 .015 .012 .16

Hardness

Two more thin samples were cut and hardness was measured in the spine, the hamon area, and the edge. Values are in Rc.

Oil Quench Water Quench

Hardness surveys, Rockwell C Hardness

A small surprise is that the oil-quenched spine is actually harder than the water quenched spine. This has been explained by the fact that the oil-quenched blade had less clay applied.

Discussion

The differences in microstructure between the oil quenched blade and the water-quenched blade are minimal. However, there are some subtleties that are worth discussion.

Spine. The spine on both blades appeared to be a simple ferrite/pearlite aggregate at low magnification. However, at higher magnification, it was clear that the dark areas in Fig. 10 and 11 were fine pearlite but the light areas were extremely difficult to resolve. Some suggestions of a shear morphology can be seen, implying that the light areas could be Bainitic ferrite or tempered martensite. The case for it being tempered martensite is made based upon the presence of temper carbides, seen as very small black dots. Regrettably, the carbides were so small that the camera did not reproduce them. However, the hardness is more consistent with very fine pearlite. White dots in the two micros were most likely cementite although there may be some ferrite at prior austenitic grain boundaries.

Hamon. The hamon (Figs. 6-9) is a mixture of two distinct structures, tempered martensite and fine pearlite. Even at 1000X, the dark areas could not be resolved sufficiently to identify the characteristic lamellar carbides in pearlite. The cooling rate in this area was fast enough to keep the inter-lamellar spacing extremely small. This microstructure would have been called troostite except that the term has fallen into disuse in recent years. It is also possible that these areas also contain bainite but again it was too fine to resolve with light microscopy. The biggest difference between the two blades was the relative amounts of pearlite and tempered martensite. The oil-quenched blade shows a higher percentage of pearlite and the water-quenched blade shows a higher percentage of tempered martensite. However, as one scans inward from the edge at the hamon, the percentages change on both blades.

A question posed here is what causes the white line above the actual hamon? It is not a clear change in the microstructure but may be related to a slight diffusion of carbon from under the clay, which may heat more slowly than the edge. Since carbon is more soluble in the hotter steel, it may diffuse toward the edge, leaving a small area under the clay slightly lower in carbon. This is only speculation, however. It may also be the way the light is reflected off of a mixed microstructure.

Edge. The edge of both blades was dense, fine, tempered martensite. Black spots in the structure may be bainite or fine pearlite. Very little difference was seen in edges of the two blades

The Bends

Evidently, the phenomenon of the blade bending down with oil quenching and bending up with water quenching has been observed for a long time. Many bladesmiths have reported the same results. Several people have speculated that the up-turned blade quenched in water is the result of the expansion of the martensite on the edge. High carbon martensite can exhibit a volumetric expansion of up to 4%. This could easily cause the edge to get longer and bow the blade, especially if the spine has not completely transformed to pearlite and some austenite remains. Explaining why the oil causes the edge to bow down is more complicated.

The most reasonable explanation has to do with the timing of the formation of the various phases. It could be that in an oil quench, the edge shrinks due to thermal contraction and the fine pearlite/bainite of the spine forms before the edge completely transforms to martensite. The formation of martensite on the edge cannot overcome the strength of the spine and the edge stays bent downward. While pearlite has a slightly greater volume per mass than does austenite, it is only slightly more, approximately .85%. It would not be likely to overcome the much greater expansion of the martensite and does not explain the downward bend.

The microstructures did not clearly indicate why the bending phenomenon happens as it does. While most of us would like to fully understand this event, perhaps the blade is not ready to expose all of the mysteries it has nurtured over the centuries.

Robert K. Nichols, PE

Consulting Metallurgical Engineer